Localization of decorin gene expression in normal human breast tissue and in benign and malignant tumors of the human breast
- First Online:
- Cite this article as:
- Boström, P., Sainio, A., Kakko, T. et al. Histochem Cell Biol (2013) 139: 161. doi:10.1007/s00418-012-1026-0
- 1.1k Downloads
The small extracellular matrix proteoglycan decorin which possesses a potent antitumor activity has been shown to be present in various amounts in the stroma of several tumors including those of the breast. Regarding decorin in breast malignancies the published data are conflicting, i.e., whether breast cancer cells express it or not. Here, we first compared decorin gene expression levels between healthy human breast tissue and selected types of human breast cancer using GeneSapiens databank. Next, we localized decorin mRNA in tissue specimen of normal human breast, intraductal breast papillomas and various histologic types of human breast cancer using in situ hybridization (ISH) with digoxigenin-labeled RNA probes for decorin. We also examined the effect of decorin transduction on the behavior of cultured human breast cancer MCF7 cells. Analysis of GeneSapiens databank revealed that in various human breast cancers decorin expression is significant. However, ISH results clearly demonstrated that human breast cancer cells independently of the type of the cancer do not express decorin mRNA. This was also true for papilloma-forming cells of the human breast. Indeed, decorin gene expression in healthy human breast tissue as well as in benign and malignant tumors of human breast was shown to take place solely in cells of the original stroma. Decorin transduction using decorin adenoviral vector in decorin-negative MCF7 cells resulted in a significant decrease in the proliferation of these cells and changed cell cohesion. Decorin-transduced MCF7 cells also exhibited increased apoptosis. In conclusion, our study shows that in human breast tissue only cells of the original stroma are capable of decorin gene expression. Our study also shows that transduction of decorin in decorin-negative human breast cancer cells markedly modulates the growth pattern of these cells.
KeywordsBreast cancerExtracellular matrixDecorinIn situ hybridizationCell behavior
In recent years, growing interest has been focused on the role and therapeutic potential of extracellular matrix (ECM) macromolecules in various diseases including the evolution and progression of cancer (Hielscher et al. 2012; Järveläinen et al. 2009; Lu et al. 2012; Marastoni et al. 2008). This is because today we understand that ECM macromolecules form not only an inert, space-filling microenvironment around the cells but they also interact with cells and generate signals that regulate the behavior of the cells (Hynes 2009; Rozario and DeSimone 2010) and control angiogenesis (Hielscher et al. 2012). Indeed, individual ECM macromolecules display important functional roles in the control of key cellular events of physiologic and pathological processes, namely adhesion, migration, proliferation, differentiation, and survival (Daley et al. 2008; Järveläinen et al. 2009; Marastoni et al. 2008; Rozario and DeSimone 2010).
An ECM molecule that has been shown to be involved in the regulation of all the aforementioned cellular events, and thereby markedly contributes to health and disease, is decorin, a small leucine-rich extracellular matrix proteoglycan (Ferdous et al. 2010; Iozzo and Schaefer 2010; Seidler and Dreier 2008). The evidence suggests that decorin represents a potent antitumor molecule (Iozzo and Sanderson 2011; Theocharis et al. 2010). For example, early studies with decorin-deficient mice have indicated that although the lack of decorin does not lead to the development of spontaneous tumors (Danielson et al. 1997), it is permissive for tumorigenesis (Iozzo et al. 1999). In concordance with this, we have recently demonstrated that there is a striking difference in the expression of decorin between malignant and benign vascular tumors in human, i.e., within Kaposi’s sarcoma, and angiosarcoma, decorin expression is completely lacking; while within hemangiomas, decorin is expressed in abundant amounts (Salomäki et al. 2008). Decorin has also been shown to inhibit tumor growth by antagonizing tumor angiogenesis (Neill et al. 2012). Furthermore, ectopic expression of decorin has been shown to cause generalized growth suppression in neoplastic cells of various histologic origin (Santra et al. 1997). As can be expected, several other studies supporting the antitumor and antimetastatic activity for decorin have been published (Biaoxue et al. 2011; Goldoni and Iozzo 2008; Hu et al. 2009; Reed et al. 2005; Shintani et al. 2008; Troup et al. 2003), and low levels of decorin have been found to be associated with a shorter progression time and poorer survival in lymph node-negative invasive human breast carcinomas (Araki et al. 2009; Troup et al. 2003). As such, a lot of interest has been paid to the potential use of decorin as an anticancer agent in the future (Neill et al. 2012; Pucci-Minafra et al. 2008; Theocharis et al. 2010).
Breast cancer is the leading cancer malignancy among women aged 20–59 years (WHO 2009). It comprises a collection of diseases that have different histopathology, genetic, and genomic variations, and prognostic outcomes (Geyer et al. 2009). In both benign and malignant breast tumors, alterations of stromal structure and composition are well-recognized phenomena (Alowami et al. 2003; Brown et al. 1999; Lu et al. 2012). These alterations in turn are likely to play an important role in the growth and invasion of breast lesions (Brown et al. 1999; Lu et al. 2012). Regarding decorin, its expression has been shown to be increased in the peritumoral stroma of the malignant lesions but decreased within the breast tumor tissue (Brown et al. 1999; Leygue et al. 2000). However, results on decorin expression in human breast cancer have been somewhat conflicting, i.e., whether breast cancer cells express this small proteoglycan or not (Cawthorn et al. 2012; Gu et al. 2010; Leygue et al. 2000; Skandalis et al. 2011). Thus, in the present study, we aimed to localize decorin mRNA in individual cells within normal and malignant human breast tissues using in situ hybridization (ISH) with digoxigenin (DIG)-labeled RNA probes for decorin. In addition, by utilizing cultures of MCF7 human breast adenocarcinoma cells and a decorin producing adenoviral vector, we also examined whether targeted decorin delivery can modulate the behavior of these cells.
Materials and methods
Patients and tumors
Patients and tumor characteristics
Number of patients (%)
Number of the patients
69 (aged 40–94, mean 65.4)
8 (11.6 %)
36 (52.2 %)
25 (36.2 %)
Axillary nodal status
33 (47.8 %)
33 (47.8 %)
3 (4.3 %)
23 (33.3 %)
46 (66.7 %)
Estrogen receptor status (ER)a
56 (81.2 %)
13 (18.8 %)
Progesterone receptor status (PR)a
53 (76.8 %)
16 (23.2 %)
Low ≤15 %
23 (33.3 %)
Intermediate 16–30 %
26 (37.7 %)
High >30 %
20 (29 %)
49 (71.0 %)
11 (15.9 %)
9 (13.0 %)
IHC positive (2+ and 3+)
27 (39.1 %)
IHC negative (0 and 1+)
42 (60.9 %)
11 (15.9 %)
The GeneSapiens database was used to analyze previous published results of the gene expression levels of decorin in healthy and malignant human breast tissues (Kilpinen et al. 2008). This database (http://www.genesapiens.org/) covers the relative gene expression patterns for 17,330 genes across all the 9,783 annotated healthy and pathological human tissue samples from publicly available Affymetrix microarray experiments. The database contains 15 healthy breast tissue samples and 1,504 different human breast carcinoma samples.
Five different ready-to-use mouse or rabbit monoclonal antibodies were used from Ventana Medical Systems/Roche Diagnostics: Estrogen Receptor (SP1, rabbit), Progesterone Receptor (1E2, rabbit), HER-2/neu (4B5, rabbit), Ki-67 (30-9, rabbit), and p63 (4A4, mouse) with BenchMark XT immunostainer and ultraVIEW Universal DAB Detection Kit (Ventana/Roche, Tucson, Arizona, USA). The percentage of nuclei with immunoreactivity to ER, PR, and Ki-67 was classified as continuous data from 0 to 100 %. ER-positive and PR-positive cases showed staining in at least 10 % of the tumor cell nuclei. Carcinomas revealing 3+ immunohistochemical membrane staining for Her2 or positive gene amplification by CISH were accepted as positive (Boström et al. 2009). Immunostaining for p63 was done to identify myoepithelial cells in some cases. Immunohistochemical analyses for decorin were performed as previously described in detail (Salomäki et al. 2008).
Decorin in situ hybridization
Decorin ISH was performed on 4 μm breast tissue sections by probing with human decorin antisense and sense single-stranded RNA riboprobes. A 533 bp fragment containing human decorin cDNA was cloned into the Eco RI/Hind III site of pGEM-4Z transcription vector (kindly provided by Dr. Liliana Schaefer, University of Frankfurt, Frankfurt am Main, Germany). Linearized plasmid DNA was purified with QIAquick PCR Purification Kit (QIAGEN, Hilden, Germany) and DIG-labeled sense and antisense RNA probes were synthesized by in vitro transcription with SP6 and T7 polymerases, respectively, using a DIG RNA Labeling Kit (Roche, Applied Science, Mannheim, Germany). Probe quantification was carried out with a DIG Nucleic Acid Detection Kit (Roche), and ISH was performed as described (Salomäki et al. 2008).
For transduction experiments, a recombinant replication-deficient adenoviral vector dcn-pxc1c-1 was used. This vector harbors the human decorin (DCN) cDNA under the control of cytomegalovirus (CMV) promoter. For the preparation of the vector, full length human decorin cDNA (Fisher et al. 1989) in pGEM plasmids was cloned and inserted into shuttle plasmid pxcJL-1. The viruses were prepared by cotransfecting HEK293–cells with back bone plasmid pBHG10. As a control vector RAdlacZ, which harbors the E. coli β-galactosidase gene (lacZ) under the control of CMV IE promoter was used (Wilkinson and Akrigg 1992). This vector was purchased from the Virus Vector Facility, Centre for Biotechnology, University of Turku, Turku, Finland.
Human breast adenocarcinoma cell line MCF7 was used for transduction with a recombinant replication-deficient adenoviral vector dcn-pxc1c-1. MCF7 cells were maintained in RPMI-1640 medium containing 10 % fetal bovine serum (FBS), 25 μM insulin, 1 nM β-estradiol, 2 mM l-glutamine, penicillin (100 IU/mL), and streptomycin (100 μg/mL) and grown at 37 °C with 5 % CO2. The cells were plated on a 24-well plate (Greiner Bio-One, Kremsmuenster, Austria), 30,000 per well. The next day, cells were transduced with 0, 3, 30, 100, 300, and 1000 pfu/cell of dcn-pxc1c-1 or RAdlacZ in reduced medium containing no FBS. Four parallels were made of each vector concentration. After 24-h incubation, the cells were washed twice with reduced medium and incubated in this medium for another 24 h. The cells were trypsinized, pooled, and the RNA was extracted using NucleoSpin RNA II–kit (Macherey–Nagel, Düren, Germany) according to the manufacturer’s instructions.
RNA concentration from the extractions was determined using a Nano-Drop spectrophotometer (ThermoScientific, Waltham, MA, USA), and the integrity of the RNA was confirmed with agarose gel electrophoresis. One μg of RNA was DNase treated with RQ1 RNase-Free DNase (Promega, Madison, WI, USA) and reverse transcribed into cDNA using M-MLV reverse transcriptase and Oligo(dT)15 primer (Promega, Madison, WI, USA) according to manufacturer’s instructions. RT-qPCR was performed using GoTaq qPCR Master Mix (Promega, Madison, WI, USA) with 100 nM primer concentrations and final volume of 10 μL according to manufacturer’s protocol. GNB2L1 was chosen as a reference gene (Zhang et al. 2005). Primer pairs used in qPCR were: 5′-GGACCGTTTCAACAGAGAGG-3′ (for) and 5′-GAGTTGTGTCAGG GGGAAGA-3′ (rev) for decorin and 5′-GAGTGTGGCCTTCTCCTCTG-3′ (for) and 5′-GCTTG CAGTTAGCCAGGTTC-3′ (rev) for GNB2L1. Reactions were run on an Applied Biosystems 7900HT machine (Applied Biosystems, Carlsbad, CA, USA). The qPCR protocol consisted of initial denaturation at 95 °C for 2 min followed by 40 cycles of denaturation at 95 °C for 40 s and extension at 60 °C for 45 s. The specificity of the reactions was confirmed by melt-curve and agarose gel analysis. Triplicate CT values were analyzed using the comparative CT (2−△△CT) method.
Unpaired Student’s t test was employed in statistical analyses. All p values <0.05 were considered statistically significant.
Relative decorin gene expression in human breast cancer tissues based on the GeneSapiens in silico transcriptomics data
Localization of decorin mRNA in normal human breast tissue, and in benign and malignant tumors of the human breast
Effect of adenoviral decorin transduction on MCF7 cells
Previously, decreased level of decorin expression has been reported in breast cancer (Eshchenko et al. 2007; Gu et al. 2010; Leygue et al. 2000). However, in a recent study, even increased amounts of decorin in breast carcinoma has been observed (Skandalis et al. 2011). In addition to these conflicting results, it has also not been convincingly demonstrated whether breast cancer cells themselves express decorin or not (Cawthorn et al. 2012; Gu et al. 2010).
In the present study, we have first analyzed previously published data on decorin gene expression using GeneSapiens databank. Thereafter, we have localized decorin mRNA in tissues samples of normal human breast and selected human breast tumors using ISH with DIG-labeled decorin probes. We have also examined the influence of adenoviral mediated decorin transduction on the behavior of human breast cancer cells in vitro. GeneSapiens databank analysis demonstrated that the relative decorin gene expression is significant in both healthy and various malignant conditions of human breast tissue. ISH of healthy human breast tissue and selected human breast tumor samples revealed that decorin mRNA can merely be localized to the normal stroma. Indeed, no decorin mRNA could be detected in benign epithelial cells or in malignant cancer cells of the human breast. In previous studies, it has been demonstrated that decorin expression is high in normal human breast tissue stroma adjacent to lobules and reduced in the breast tumor itself (Brown et al. 1999; Leygue et al. 2000). However, these studies have used radioactive probes use of which is limited, if the purpose is to exactly localize the mRNA at the cellular level. With radioactive probes, all the cellular relationships are lost because of tissue digest and mRNA levels are averaged from all of the cells contained in the original sample (Wilcox 1993). When methods like Northern (Tralhão et al. 2003) or western blot analyses (Leygue et al. 2000; Nash et al. 2002; Troup et al. 2003) and more modern methods such as RT-PCR (Eshchenko et al. 2007) are used to evaluate the expression of molecules, e.g., decorin, they are usually performed with pooled samples consisting of the sectioned tumor which also contain its surrounding original stroma. Therefore, these above-mentioned methods including GeneSapiens database used in the present study are unable to tell the cellular origin of a specific molecule. Our results using ISH with DIG-labeled decorin probe have clearly shown that human breast cancer cells do not express decorin at all and that in human breast tissue specimen decorin is derived from original stromal cells, not from benign or malignant tumor-forming epithelial cells. However, the specific cell types responsible for decorin expression in the original stroma compartment still remains to be studied. To the best of our knowledge, this study is the first to exactly localize decorin mRNA at the cellular level in human breast tissue including different types of human breast cancer.
Next, we examined whether by introducing decorin, a known antitumoral molecule, to widely used MCF7 (Burdall et al. 2003) human breast adenocarcinoma cells, we could modulate the behavior of these cells. We showed that decorin transduction causes marked changes in the proliferation and growth pattern of the MCF7 cells. In particular, compared to control MCF7 cells, decorin-transduced MCF7 cells exhibited statistically significantly lower mitosis rate, and they also revealed increased apoptotic features such as the formation of the so-called cannibal cells. Earlier, it has been shown in mice that decorin gene delivery decreases the proliferative index (Pucci-Minafra et al. 2008) and alters the architecture and differentiation of the tumor xenografts, thereby retarding the growth of colon and squamous cell carcinomas (Reed et al. 2002). Using an animal model of orthotopic breast carcinoma, decorin has also been shown to slow mammary carcinoma cell motility, induce significant apoptosis and impede cell invasion through a three-dimensional extracellular matrix (Goldoni et al. 2008).
The mechanism(s) of action of decorin transduction on the behavior of MCF7 human breast cancer cells was not in our focus in this study and therefore remains to be explored. However, the observed decreased cell cohesion of MCF7 cells in response to decorin transduction could be caused by modulated pericellular matrix around the cells, because previously decorin has been shown to influence cellular adhesion via its capability to interact e.g., with fibronectin (Winnemöller et al. 1991) and thrombospondin (Winnemöller et al. 1992). It is also possible that alterations in the expression of metalloproteinases (MMPs) and their inhibitors (TIMPs) by MCF7 cells are responsible for the observed decreased cell cohesion. Indeed, adenovirus-mediated overexpression of decorin has been shown to modulate the expression of certain MMPs and TIMPs by human gingival fibroblasts (Al Haj Zen et al. 2003). Furthermore, the possibility that the decreased cell cohesion of MCF7 cells in response to decorin transduction is mediated via direct interactions between decorin and certain integrins remains to be explored (Guidetti et al. 2002). The decreased proliferation of decorin-transduced MCF7 cells, on the other hand, is most likely mediated via decorin’s ability to antagonize receptor tyrosine kinases, particularly the signaling through epidermal growth factor receptor (Hu et al. 2009), the receptor for hepatocyte growth factor called Met (Goldoni et al. 2009), insulin-like growth factor receptor (Iozzo et al. 2011) and/or vascular endothelial growth factor 2 receptor (Khan et al. 2011). In addition, induction of apoptotic features of MCF7 cells due to decorin transduction may also have a role in the finding of this study that decorin-transduced MCF7 human breast cancer cells exhibit decreased proliferation. Interestingly, exogenous decorin core protein has been shown to inhibit cancer growth by triggering apoptosis via activation of caspase-3 (Seidler et al. 2006). Finally, decorin’s ability to bind and block transforming growth factor-β should be kept in mind (Yamaguchi et al. 1990), although transforming growth factors have been shown not to act as major growth regulators of MCF7 cells (Herman and Katzenellenbogen 1994).
In conclusion, by utilizing ISH with DIG-labeled probes for decorin, we have shown that in human breast cancer, the cancer cells do not express decorin mRNA, but the expression of decorin takes place merely in cells of the original stroma both in healthy human breast tissue and in breast tissues containing benign or malignant breast epithelial tumors. By introducing a decorin producing adenoviral vector to the MCF7 human breast adenocarcinoma cells, we have also shown that decorin transduction in these decorin-negative cells results in marked changes in their behavior. Specifically, we have shown that decorin-transduced MCF7 cells exhibit decreased cell cohesion and statistically significantly lower mitosis rate and increased apoptotic features. Thus, our study provides evidence that targeted decorin transduction to breast cancer cells could be used as a novel adjuvant therapy in the treatment of human breast cancer in the future.
The authors are grateful to Ms. Minnamaija Lintunen, PhD, for the validation of the immunohistochemical stainings. The expert technical assistance of Ms. Sinikka Kollanus is gratefully acknowledged. Ms. Henriikka Salomäki, MSc, is also gratefully acknowledged for her excellent technical assistance during the early stages of this study. Financially, this study was supported by Medical Research Fund (EVO) of Turku University Hospital, Cancer Foundation of South-Western Finland, and by grant from the Finnish Breast Cancer Group.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.